Trial Information

Development and Evaluation of Techniques for Computer Aided Detection and Diagnosis From Radiologic Images

Radiologic images are becoming more and more complex, and utilization of radiologic
techniques is accelerating. Radiologists and other clinicians are being inundated with
radiologic data. Computer aided detection and diagnosis (CAD) have the potential to improve
patient care by increasing sensitivity of diagnostic tests, reducing false positives and
improving physician efficiency. Computer aided detection and diagnosis have been under
development for many years yet there is still much work to be done to move it from the bench
to the bedside. The purpose of this project is to develop and evaluate techniques for CAD
using the existing radiologic data available in the Clinical Center's Department of
Diagnostic Radiology. Such techniques include but are not limited to automated detection of
melanoma, bone metastases and pulmonary emboli, colonic polyps and abdominal tumors, and
abnormal morphology of organs. The outcome of this study will be algorithms and software
that accurately detect and characterize lesions, model anatomy and monitor diseases on
radiologic studies.